The best data scientists are critical thinkers par excellence. Acquiring skills in quantitative analysis and programming is not enough, and even a strong background in some domain specialty doesn’t guarantee a data scientist’s effectiveness. Data scientists also need the ability to drill conceptually through the heart of whatever problem they’re analyzing. Thinking like a data scientist is no mean feat. It may require thinking outside the box, and even beyond the scope of the current data set and constructed statistical model, in the quest for the golden insight.
In IBM Data magazine the week of January 26, 2015, we have three excellent new articles focusing on different types of thinking that are the key to productivity as a data scientist. People’s words and actions frequently disconnect from each other, and behavioral analytics is often a necessary counterweight to linguistic analysis when data scientists seek the heart of intention and propensity. First-time contributor Andy Thurai shows how Match.com’s data scientists tackled a thorny analytics problem by rethinking what data sources are most relevant to customer behaviors. “Specifically,” Thurai writes, “Match.com figured out that building models based on what people say—their wants—is not enough. In other words, it determined that users’ actions and their actual needs are totally different from their wants. Ultimately, Match.com was able to predict user behavior based on users’ actions on its Match.com website with an enhanced success rate.”
Telling factual stories with data analytics is where the best data scientists shine. Rich Hughes provides a comprehensive profile of what it takes to be an effective data scientist. And much of what it takes hinges on not only being able to craft compelling data visualizations, but also using them to explain their findings to colleagues, customers, and other stakeholders and interested parties. Thinking skills include the thoughtfulness of data scientists who are sensitive to the full social context of their efforts. Hughes cites a data scientist’s astute observation: “‘People make a mistake by forgetting that data science is a team sport.… [T]here’s not one single data scientist that does it all on [his or her] own.’”
Data scientists need to think like detectives. The data discovery process often involves sleuthing for clues regarding what data may be available and which of it is most relevant to the problem at hand. Edd Dumbill takes us inside the data discovery value chain, spelling out the diverse criteria that working data scientists use to sift through data sets of any volume, variety, and velocity. Just as important, he says, data scientists must grin and bear the tedium that such searches often entail. “In the data science world,” Dumbill writes, “obtaining and cleaning data is often recognized as always being 80 percent of the work of any analysis, because of database silos within organizations and a complex and opaque ecosystem of external data sources.”
Thanks for reading and engaging. And please check out our latest NewsBytes and upcoming events for opportunities to educate yourself on the power of data.
James Kobielus (@jameskobielus)
Editor in Chief, IBM Data magazine
The best data scientists are critical thinkers par excellence. Acquiring skills in quantitative analysis and programming is not enough, and even a strong background in some domain specialty doesn’t guarantee a data scientist’s effectiveness.